Fig. 1 Fig. 2

Early-onset idiopathic scoliosis-a very serious condition producingcardiopulmonary compromise and horrendous deformity. Figure 1-

Erect. Figure 2-Close-up in the forward-bending position.

Fig. 3 Fig. 4

Late-onset idiopathic scoliosis-a problem of deformity only.Figure 3-Erect. Figure 4-Close-up in the forward-bending position.

REVIEW ARTICLE

I 76 THE JOURNAL OF BONE AND JOINT SURGERY

CONSERVATIVE TREATMENT FOR IDIOPATHIC SCOLIOSIS

R. A. DICKSON

Because idiopathic scoliosis commences and may pro-

gress during the period of spinal growth, it has been

subdivided according to when it begins (James 1954);

thus infantile, juvenile and adolescent types are rec-

ognised (Goldstein and Waugh 1973). While a one-

year-old baby with a 60 idiopathic thoracic curve

unquestionably has infantile scoliosis, classification

becomes progressively more difficult the older the child;

and although a 12-year-old girl with a 90 curve is an

adolescent, she does not have adolescent-onset

idiopathic scoliosis, as the deformity certainly started

many years earlier. Furthermore, there is no clear

evidence that juvenile-onset idiopathic scoliosis exists. Of

James’ 134 patients with thoracic scoliosis, only I 6 were

tentatively classified as ofjuvenile onset and he did not

think these worth separating from the infantile group

(James 1954). Such cases may well be a hangover from

infancy (Mehta 1977). There is much merit therefore in

considering only two categories, early-onset and late-

onset (Ponseti and Friedman 1950; Figs I to 4). The

prevalence rate, natural history and the consequences of

untreated scoliosis, as well as the strategy for treat-

ment and its efficacy, differ very considerably between

early and late-onset types. Treatment for the more

common and more benign late-onset case is more

“standard” and will therefore be discussed first.

LATE-ONSET IDIOPATHIC SCOLIOSIS

The need for treatment. The strategy for treating

idiopathic scoliosis depends principally upon the size of

the deformity and its potential for progression. If the

deformity is acceptable at presentation, then preservation

of acceptability is the aim; this is the place of conserva-

tive management. If the deformity is unacceptable, then

the objective must be to make it acceptable and keep it

so; this is the aim of surgical management. In order to

decide if idiopathic scoliosis needs treatment at all, the

consequences of leaving it untreated must be known. It

is, of course, known that scoliosis can cause significant

disability, with economic implications (Dahlberg and

R. A. Dickson, MA, ChM. FRCS, Professor and Head of theDepartment ofOrthopaedic and Traumatic Surgery, The University ofLeedsSt Jarness University Hospital, Leeds LS9 7TF, England.

(� 1985 British Editorial Society of Bone and Joint Surgery

030l-620X 85/2013 S2.00

CONSERVATIVE TREATMENT FOR IDIOPATHIC SCOLIOSIS 177

VoL. 67-B. No. 2. MARCH 1985

Nachemson 1977), and that the cardiopulmonary com-

plications can be a source of morbidity and mortality

(Nachemson 1968; Nilsonne and Lundgren 1968; Collis

and Ponseti 1969). But this applies only to early-onset

cases; and then only to severe ones, for example, a

thoracic curve of over 60 by the age of seven or eight

years. when the pulmonary parenchyma is developing

(Davies and Reid 1971).

In contradistinction there are no obvious organic

consequences of late-onset idiopathic scoliosis, even if

the deformity exceeds 100 (Kostuik, Israel and Hall

1973; Ponder ci al. 1975; Dickson and Leatherman

1976). Late-onset idiopathic scoliosis is a problem of

deformity only. The 52 late-onset idiopathic cases in

Nachemson’s original study of 130 patients fared no

differently from their straight-backed counterparts as

regards organic health (Nachemson 1968). With regard

to the deformity, it is of course true that the bigger the

deformity the greater the likelihood of social and

psychological implications (Nilsonne and Lundgren

1968; Bengtsson ci’ al. 1974). The patient’s opinion on

this subject clearly matters more than that of the

surgeon.

Natural history. As there has never been a controlled trial

of conservative treatment its efficacy can only be deter-

mined by evaluating it against the little we know of the

natural history of the late-onset curve. There are two

quite different sources of information. Early studies of

selected groups of children who presented to scoliosis

clinics suggested a considerable progression potential if

the onset was under 10 years of age or before the

menarche (Risser and Ferguson 1936; Ponseti and

Friedman 1950; James 1954; Heine and Reher 1975).

More recent data come from those school screening

programmes which have included a longitudinal survey

(Brooks et al. 1975; Rogala, Drummond and Gurr 1978;

Dickson et al. 1980; Dickson 1983). When children with

non-structural curves are excluded, only 10% show

evidence of progression, whereas twice as many improve

and more than two-thirds remain static. The greatest

progression potential is associated with the young girl

who has a right thoracic curve, but she represents less

than one in a thousand of those screened. The difference

between the data from these two sources, although

difficult to interpret (Leaver, Alvik and Warren 1982),

suggests a change to a more benign natural history, and

this is supported by the observation that where a lot of

screening has been performed the need for both conser-

vative and operative management has been much less

(Lonstein el a!. 1982). Conservative treatment should

then be set against this background. Three conservative

methods need to be discussed: bracing, casting and

electrospinal stimulation.

Brace treatment. While various contraptions for the

scoliotic spine have been used since the time of

Hippocrates. real enthusiasm for conservative treatment

started with the Milwaukee brace (Blount and Schmidt

1957; Blount 1958). Newer technology refers to this

brace as a CTLSO (cervical-thoracic-lumbar-sacral or-

thosis) (Nash 1980). Although the brace was not pri-

manly intended for the conservative management of

idiopathic scoliosis, it was soon used for that purpose

(Blount 1972; Moe 1973). Early mechanical studies

suggested that the brace might function by exerting

distraction between the head and the pelvis (Schultz and

Galante 1969; Galante et al. 1970) and this mode�of

action was corroborated by the harmful effects on

dentition thereby produced (Alexander 1966). A change

to the throat mould type ofbrace led to a great reduction

in the distraction force with no obvious dental problems

(Northway, Alexander and Riolo 1974); and a change in

biomechanical approach led to three-point fixation with

particular emphasis placed on the localiser pad

(Andriacchi ci al. 1976).

Without a clear understanding of the three-

dimensional nature of the deformity it would be tempt-

ing to think that the brace might work in the manner

described, but this is not so. The primary deformity of

idiopathic scoliosis is a lordosis at the curve apex

(Adams 1882; Somerville 1952; Roaf 1966; Dickson ci al.

1983, 1984) and it is rotation of this lordosis to the side

which produces the secondary scoliotic deformity. An

ideal ofconservative treatment would then be to recreate

the normal spinal shape in the sagittal plane; this,

however, would imply flexion, which enhances rotation

and produces an increase in the secondary scoliotic

deformity. In contradistinction, the opposite deformity,

the kyphosis of Scheuermann’s disease, is ideally suited

to conservative management, because the deformity is

rotationally stable and braces which cause spinal exten-

sion produce a true physiological correction of the

deformity (Bradford ci al. 1974). The brace is capable,

however, of producing a small temporary corrective

effect in idiopathic scoliosis. Blount stressed the need for

obliteration of the lumbar lordosis in the brace (Blount

and Moe 1973) and this produces thoracic extension

above. There is now more room for the thoracic lordosis

to be accommodated with a derotation effect, but at the

possible expense of increasing the primary lordosis

(Winter, Lovell and Moe 1975; Figs 5 and 6). The other

important effect of the brace is to splint the spine to the

pelvis in the erect position which thus prevents the

harmful effect of flexion.

With this mode of action it is not surprising to find

that the optimal result of brace wearing is when the curve

measures exactly the same at the end of treatment as it

did at the beginning (Keiser and Shuffiebarger 1976;

Edmondson and Morris 1977; Mellencamp, Blount and

Anderson 1977; Tolo and Gillespie 1978: Blount 1981).

These studies also suggest that the more the curve has

progressed beyond 30 before the commencement of

treatment, the less satisfactorily can curve progression be

attenuated, as gravity and the rigidity of the secondary

deformities more successfully defeat the intentions of

Fig. 5

178 R. A. DICKSON

THE JOURNAL OF BONE AND JOINT SURGERY

Fig. 6

The effect of obliteration of the lumbar lordosis. Figure 5-A thoracicdeformity associated with a lumbar lordosis viewed from above.Figure 6-The same deformity viewed from above with the lumbar

lordosis obliterated. Marked derotation has occurred.

treatment. The best results of brace treatment are

therefore achieved with smaller curves; their natural

history, however, demonstrates that very few would have

progressed if left untreated.

The brace is supposed to be worn for 23 hours out of

24, although it would appear to be unnecessary at night

when repeated cycles of spinal flexion are not usually

performed. Although exercises have no corrective effect

on idiopathic curves (Stone ci a!. 1979), an exercise

programme is prescribed for the one hour a day spent

out of the brace lest the spine become unduly stiff; this

programme should not, however, include flexion exer-

cises as these will undo what the brace has been trying to

achieve for the previous 23 hours.

Set against the background of natural history there

is no evidence that Milwaukee brace treatment alters the

course of the scoliosis. This is a very serious matter, as

countless numbers of children may have endured brace

treatment for no detectable benefit. If the effect of the

brace on girls with progressive thoracic curves was not

diluted by the inclusion ofcurves at other sites with little

or no progression potential, then the lumbar lordosis-

obliterating and flexion-preventing effect of the brace

ought to prevent progression (Dickson ci a!. 1984).

Cognisant of this problem the British Orthopaedic

Association and the British Scoliosis Society (1983) are

right to stress the need for carefully controlled studies of

idiopathic scoliosis, preferably throughout life.

It recently became apparent that low thoracic and

lumbar curves did not require the full superstructure of a

Milwaukee brace (Park ci a!. 1977; Watts, Hall and

Stanish 1977; Winter and Carlson 1977); this is because

flexion of these low curves can be prevented even by an

underarm brace, TLSO (thoracic-lumbar-sacral ortho-

sis). But the other mode of action, obliteration of thelumbar lordosis, is more obvious and produces a bigger

temporary corrective effect. There are no controlled

trials with an underarm brace and the follow-up is much

shorter than with a Milwaukee brace; consequently their

efficacy also is questionable, though of course their

lighter weight and smaller size make them more accept-

able to the patient.

Another problem concerns the duration of spinal

growth. The only period after the intra-uterine phase

when growth velocity increases is during the adolescent

growth spurt, which is maximal at about the age of 12

years in girls and 14 in boys (Scammon 1927). While

idiopathic curves are particularly liable to deteriorate

during this phase, general skeletal maturity is reached two

years later (Tanner 1962). The conventional time when

the patient is weaned from the brace has been when the

iliac crest and vertebral ring apophyses fuse (Risser and

Ferguson 1936; James 1954; Risser 1964). It is well

known, however, that spinal growth continues for a

further 10 years until the vertebral epiphyses are fused

and that the vertebral apophyses have nothing whatever

to do with spinal growth nor does their fusion indicate

cessation ofgrowth (Bick, Copel and Spector 1950; Bick

and Copel I 95 1 ; Inkster I 95 1 ; Calvo 1957; Tupman I 962;

Larsen and Nordentoft 1962; Bernick and Caillet 1982).

Recent studies of idiopathic curves beyond general skel-

etal maturity do, in fact. demonstrate progression in the

majority ofcases (Hassan and Bjerkreim 1983; Weinstein

and Ponseti 1983). While these studies have suggested

that the effect of pregnancy on ligaments might be a

responsible factor, it ought not to be forgotten that, in

young women, the spine is still growing. Even if there

was evidence that the brace did prevent progression,

treatment would need to be continued for much longer

than the patient would tolerate.

Cast management. The pioneers of the treatment ofscoliosis obtained correction by using plaster casts

(Risser ci al. 1953; Risser 1955) and it was remarkable to

what good use they put them (Moe and Valuska 1966).

With the advent of the Milwaukee brace, enthusiasm for

plaster in the conservative treatment rapidly waned in

many parts of the world. French surgeons, however, did

not lose their faith in plaster techniques and have

developed the EDF (elongation-derotation-flexion) cast

as an alternative to brace treatment (Cotrel and Morel

1964). The function of this cast is precisely that of the

brace, with obliteration of the lumbar lordosis and

elimination of spinal flexion as the two priorities. Each

cast is worn for three or four months until its wear or the

patient’s growth indicates that a new one is required. The

patients cannot bathe, but it is extraordinary how easy it

is to change the inner vest and the underwear while the

CONSERVATIVE TREATMENT FOR IDIOPATHIC SCOLIOSIS 179

VOL. 67-B, No. 2. MARCH 1985

cast is in place. Furthermore, the cast has a window on

the concave side posteriorly and one on the convex side

anteriorly; these facilitate derotation exercises and allow

pressure pads or balloons to be inserted over the

rotational prominences. Since the object of conservative

treatment is to finish up with the least deformed torso,

“bracers” would do well to compare their end-results

with French “casters” although, as with the majority of

interesting questions, the answer has never been

elucidated by a controlled study. Between casts a pro-

gramme of non-skeletal traction and exercises is carried

out (Cotrel and D’Amore 1968). This is of no benefit,

however, in terms of curve correction (Nachemson and

Nordwell 1977; Dickson and Leatherman 1978). Indeed,

traction ofany kind provides no real correction of curves

of any magnitude, only moving each curve through

its natural range of flexibility (Edgar, Chapman and

Glasgow 1982).

Electrospinal stimulation. Recently, attention has been

directed towards obtaining temporary correction of the

scoliosis by electrical stimulation of the spinal muscula-

ture on the convexity of the curve (Bobechko 1974). Like

the development oforthotic and cast treatment, electrical

stimulation focuses on the secondary coronal-plane

deformity and only moves the spine within its natural

range of elasticity. Furthermore, electrical stimulation

stemmed from the belief that there was a neuromuscular

basis to the deformity, a belief which is unlikely to be

substantiated (Dickson ci a!. 1984).

The fact that mild coronal-plane curvatures in

animals can be produced by stimulating muscles on one

side (Olsen ci a!. 1975; Monticelli ci a!. 1975; Bobechko,

Herbert and Friedman 1976) is not surprising. This is

precisely what happens when someone with a straight

spine bends to one side and then resumes the erect

position. Some improvement in curve magnitude has

been demonstrated during convex muscle stimulation

(Bobechko, Herbert and Friedman 1979); this demon-

strates the innate flexibility that the mild idiopathic curve

enjoys, but there is no evidence that any real correction

follows electrospinal stimulation (Axelgaard and Brown

1983). Again it is the rotationally unstable nature of the

primary lordotic deformity which militates against effec-

tive conservative treatment; with the uniplanar and

rotationally stable kyphotic deformity, however, elec-

trical surface stimulation can, like the brace or cast, give

rise to permanent correction (Axelgaard, Brown and

Swank 1982).

EARLY-ONSET IDIOPATHIC SCOLIOSIS

Natural history. This fascinating condition, first reportedfrom Holland (Harrenstein 1929), tends to affect children

from birth to three years of age. Boys are affected more

commonly than girls and thoracic curves are more

frequently convex to the left (James 1951; James, Lloyd-

Roberts and Pilcher 1959; Lloyd-Roberts and Pilcher

1965: Wynne-Davies 1975; Thompson and Bentley

1980). Here progression potential is particularly relevant.

It was first thought that the condition could be divided

into two types-progressive and resolving-depending

upon the size of the rib-vertebra angle difference(RVAD; Mehta 1972). The picture is not so clear,

however, and three types are now recognised-

progressive, static and resolving (Mehta 1977). While an

RVAD ofless than 20 which then reduces in magnitude

confidently diagnoses the resolving curve, an RVAD in

excess of 20� or one that is increasing does not necess-

arily imply a progressive curve. Other factors also are

important. Thoracic and thoracolumbar curves and

small initial curves (Mehta 1977; Thompson and Bentley

1980) tend to resolve, while double structural curves havea definite progression potential (Ceballos ci a!. 1980).

The most serious progression appears to occur in thehypotonic, low birth weight baby in whom the condition

has been referred to as “malignant” idiopathic scoliosis

(Mehta 1977).

A very interesting trend has emerged over the last 35

years. Early reports indicated a great preponderance of

the progressive type of curve (James 195 1 ; Scott and

Morgan 1955; James ci a!. 1959), but this situation then

changed dramatically and the last 20 years has seen a

marked reversal of the proportions with 90% or more

resolving (Lloyd-Roberts and Pilcher 1965; Mau 1968;

Thompson and Bentley 1980; Ceballos ci a!. 1980). The

incidence of these infantile curves also has rapidly

declined and the condition is now rare; whether or not

this decline is due to prone lying in the cot is unclear

(McMaster 1983). These changes in the natural history

of early-onset progressive idiopathic scoliosis are very

welcome, as these are the curves associated with serious

cardiopulmonary disease at an early age, and they also

develop horrifying deformities.

Conservative treatment. When malignant progressive

curves were more common, treatment presented great

problems. Progression potential was far too great to be

attenuated by a Milwaukee brace (James ci a!. 1959), but

posterior fusion was withheld for as long as possible in

order to avoid increasing the primary lordosis; mean-

while the deformity progressed inexorably in the brace. By

the time posterior fusion was performed, the deformity

was often too far advanced for treatment; moreover,

there is no clear evidence that fusion reduced the rate of

subsequent progression (Letts and Bobechko 1974;

McMaster and Macnicol 1979).

Unlike late-onset deformities there is some evidence

that early-onset idiopathic scoliosis can be treated con-

servatively. Mehta, who has contributed much to our

knowledge of infantile idiopathic scoliosis, recognised

early the bad prognosis associated with the hypotonic

infant, and the moment she saw such a child she applied

an elongation-rotation-flexion (EDF) cast (Mehta and

Morel 1979). Surprisingly. the occasional case that had

all the ingredients for rapid progression appeared to

become static, or even to resolve, and the RVAD became

180 R. A. DICKSON

THE JOURNAL OF BONE AND JOINT SURGERY

smaller or did not increase. This perhaps demonstrates

the effect of obliteration of the lumbar lordosis and the

prevention of flexion in these very supple spines; but the

cast must also have allowed the thoracic spine to become

naturally kyphotic in those that subsequently resolved.

There are, clearly, two important aspects of the con-

servative management of the early-onset case, namely,

prevention and casting. Although debate continues as

to whether the deformity is due to intra-uterine moulding

(Browne 1936) or to positioning in the cot (Mau 1968),

prone lying does appear to have an inhibitory effect; it

must be insisted upon, particularly for the hypotonic

infant (McMaster 1983). For the rare case that does

develop an idiopathic curve with all the hallmarks of

progression, serial EDF casts should be applied without

delay (Mehta 1977).

REFERENCES

Adams W. Lectures on the pathology and treatment oflateral and other

forms of curvature of the spine. London: J & A Churchill & Sons,1882.

Alexander RG. The effects on tooth position and maxillofacial verticalgrowth during treatment of scoliosis with the Milwaukee brace.Ani J Orthod 1966:52: 161-89.

Andriacchl TP, Schultz AB, Belytschko TB, DeWald RL. Milwaukeebrace correction of idiopathic scoliosis: a biomechanical analysisand a retrospective study. J Bone Joint Surg [Am] 1976:58-A:806-IS.

Axelgaard J, Brown JC. Lateral electrical surface stimulation for thetreatment of progressive idiopathic scoliosis. Spine 1983;8(3):242-60.

Axelgaard J, Brown JC, Swank SM. Kyphosis treatment by electricalsurface stimulation. Orthop Trans 1982:6:1.

Bengtsson G, Fillstr#{246}mK, Jansson B, Nachemson A. A psychologicaland psychiatric investigation of the adjustment of female scoliosispatients. Acta Psrchiatr Scand l974;5O:50-9.

Bernick S. Caillet R. Vertebral end-plate changes with aging of thehuman vertebrae. Spine 1982;7(2):97-102.

Bick EM, Copel JW. The ring apophysis of the human vertebra.Contribution to human osteogeny. II. J Bone Joint Surg [Am]1951 :33-A:783-7.

Bick EM, Copel JW, Spector S. Longitudinal growth of the humanvertebra: a contribution to human osteogeny. J Bone Joint Surg[Am] 1950;32-A:803-l4.

Blount WP. Scoliosis and the Milwaukee brace. Bull Hosp Joint Dis1958:19: 152-65.

Blount WP. Use of the Milwaukee brace. Orthop Clin North Am.l972;3:3-16.

Blount WP. Editorial. The virtue of early treatment of idiopathicscoliosis. J Bone Joint Surg [Am] 198 I:63-A:335-6.

Blount WP, Moe JH. The Milwaukee brace. Baltimore: Williams &Wilkins, 1973.

Blount WP, Schmidt AC. The Milwaukee brace in the treatment ofscoliosis. J Bom’ Joint Surg [Am] I957;39-A:693.

Bobechko WP. Scoliosis spinal pacemakers. J Bone Joint Surg [Am]l974;56-A:442.

Bobechko WP, Herbert M, Friedman H. Electro-spinal instrumenta-tion. J Bone Joint Surg [Am] 1976:58-A: 156.

Bobechko WP, Herbert M, Friedman HG. Electrospinal instrumenta-tion for scoliosis: current status. Orthop Cli,i North Am1979: 1O(4):927-41.

Bradford DS, Moe JH, Montalvo FJ, Winter RB. Scheuermann’skyphosis and roundback deformity: results of Milwaukee bracetreatment. J Bone Joint Surg [Am] l974;56-A:740-58.

British Orthopaedic Association and the British Scoliosis Society. Schoolscreening for scoliosis. Brit Med J 1983:287:963-4.

Brooks HL, Azen SP, Gerberg E, Brooks R, Chan L. Scoliosis: aprospective epidemiological study. J Bone Joint Surg [Am]1975 :57-A :968-72.

Browne D. Congenital deformities of mechanical origin. Proc R SocMed I936:29: 1409-31.

Calvo IJ. Observations on the growth of the female adolescent spineand its relation to scoliosis. Clin Orthop 1957:10:40-7.

Ceballos T, Ferrer-Torrelles M, Castillo F, Fernandez-Paredes E.Prognosis in infantile idiopathic scoliosis. J Bone Joint Surg [Am]1980:62-A: 863-75.

Collis DK, Ponseti IV. Long-term follow-up of patients with idiopathicscoliosis not treated surgically. J Bone Joint Surg [Am] 1969:51-A:425-45.

Cotrel �‘, D’Amore M. Spinal traction in scoliosis. In: Zorab PA, ed.Proceedings of a second svniposium on scoliosis: causation.Edinburgh and London: E & S Livingstone. 1968:37-43.

Cotrel Y, Morel G. La technique de I’E.D.F dans Ia correction desscolioses. Ret’ Chir Orthop 1964:50:59-75.

Dahlberg L, Nachemson AL. The economic aspects of scoliosis treat-ment. In: Zorab PA, ed. Scoliosis: proceedings ofafif)h symposiumheld at the Cardiothoracic Institute, Brompton Hospital. London, on2/ct, 22nd September. 1976. London: Academic Press,1977:73-101.

Davies C, Reid L. Effect of scoliosis on growth of alveoli andpulmonary arteries and on right ventricle. Arch Dis Child1971 :46:623-32.

Dickson RA. Scoliosis in the community. Br Med J 1983:286:615-8.

Dickson RA, Lawton JO, Archer IA, Butt WP. The pathogenesis ofidiopathic scoliosis. Biplanar spinal asymmetry. J Bone Joint Surg[Br] 1984:66-B:8-l5.

Dickson RA, Lawton JO, Archer IA, et a!. Combined median andcoronal plane asymmetry: the essential lesion of idiopathicscoliosis. J Bone Joint Surg [Br] 1983:65-B:368.

Dickson RA, Leatherman KD. Spinal deformity in adults: changingconcepts. J Bone Joint Surg [Am] l976:58-A:729.

Dickson RA. Leatherman KD. Cotrel traction, exercises, casting in thetreatment of idiopathic scoliosis: a pilot study and prospectiverandomized controlled clinical trial. Aeta Orthop Scand1978 :49:46-8.

Dickson RA, Stamper P, Sharp A-M, Harker P. School screening forscoliosis: cohort study of clinical course. Br Med J 1980:281:265-7.

Edgar MA, Chapman RH, Glasgow MMS. Pre-operative correction inadolescent idiopathic scoliosis. J Bone Joint Surg [Br] 1982:64-B: 530-5.

Edmondson AS, Morris JT. Follow-up study of Milwaukee bracetreatment in patients with idiopathic scoliosis. Clin Ort hop1977:126:58-61.

Galante J, Schultz A, DeWald RL, Ray RD. Forces acting in theMilwaukee brace on patients undergoing treatment for idiopathicscoliosis. J Bone Joint Surg [Am] l970:52-A:498-506.

Goldstein LA, Waugh TR. Classification and terminology of scoliosis.Cliii Orthop 1973:93: 10-22.

Harrenstein RJ. Die Skoliose bei S#{228}uglingen und ihre Behandlung.z Orthop Chir 1929:52: 1-40.

Hassan 1, Bjerkreim I. Progression in idiopathic scoliosis after conser-vative treatment. Acta Orthop Scand 1983:54:88-90.

Heine J, Reher H. Die Progredienz der unbehandelten idiopath-ischen Skoliose bis Wachstumsabschluss. 7 Orthop 1975:113:87-96.

Inkster RG. Osteology. In: Brash JC, ed. Cunningham’s text-hook ofanatomy, 9th edn. London: Oxford University Press.195 1: 105-331.

James JIP. Two curve patterns in idiopathic structural scoliosis. J BoneJoint Surg [Br] 1951 :33-B:399-406.

James JIP. Idiopathic scoliosis: the prognosis. diagnosis and operativeindications related to curve patterns and the age at onset. J BoneJoint Surg [Br] 1954:36-B:36-49.

James JIP, Lloyd-Roberts GC, Pilcher MF. Infantile structuralscoliosis. J Bone Joint Surg [Br] 1959:41-B:719-35.

Keiser RP, Shufflebarger HL. The Milwaukee brace in idiopathicscoliosis: evaluation of 123 completed cases. Cliii Orthop1976:118: 19-24.

Kostuik JP, Israel J, Hall JE. Scoliosis surgery in adults. Clin Orthop1973 :93: 225-34.

Larsen EH, Nordentoft EL. Growth ofthe epiphyses and vertebra. ActaOrthop Sca,icl I 962:32:210-7.

Leaver JM, Alvik A, Warren MD. Prescriptive screening for adolescentidiopathic scoliosis: a review of the evidence. mt J Epidemiol1982:11(2): 101-11.

CONSERVATIVE TREATMENT FOR IDIOPATHIC SCOLIOSIS 181

VOL. 67-B. No. 2. MARCH 1985

Letts RM, Bobechko WP. Fusion of the scoliotic spine in youngchildren: effects on prognosis and growth. Clin Ort hop1974:101: 136-45.

Lloyd-Roberts GC, Pitcher MF. Structural idiopathic scoliosis ininfancy: a study of the natural history of I 00 patients. J Bone JointSurg [Br] 1965:47-B:52&-3.

Lonstein JE, Bjorklund 5, Wanninger MH, Nelson RP. Voluntaryschool screening for scoliosis in Minnesota. J Bone Joint Surg [Am]1982 :64-A : 48 1-8.

McMaster MJ. Infantile idiopathic scoliosis: can it be prevented? JBone Joint Surg [Br] l983:65-B:6I2-7.

McMaster MJ, Macnicol MF. The management of progressive infan-tile idiopathic scoliosis. J Bone Joint Surg [Br] l979:6I-B:36-42.

Mau H. Does infantile scoliosis require treatment? J Bone Joint Surg[Br] l968:50-B:88I.

Mehta MH. The rib-vertebra angle in the early diagnosis betweenresolving and progressive infantile scoliosis. J Bone Joint Surg [Br]1972:54-B: 230-43.

Mehta MH. The natural history of infantile idiopathic scoliosis. In:Zorab PA, ed. Scoliosis: proceedings of a fifth .n’mposium held atthe Cardiothoracic Institute, Brompton Hospital. London, on 21st,22nd September. 1976. London: Academic Press, 1977 : 103-22.

Mehta MH, Morel G. The non-operative treatment of infantileidiopathic scoliosis. In: Zorab PA. Siegler D. eds. Scoliosis 1979.Based on the proceedings of the sixth symposium on scoliosis heldat the Cardiothoracic Institute, Brompton Hospital, London, onSeptember 17 and 18. 1979. London: Academic Press, 1980:71-84.

Mellencamp DD, Blount WP, Anderson AJ. Milwaukee brace treatmentof idiopathic scoliosis: late results. Clin Orthop 1977; 126:47-57.

Moe JH. Indications for Milwaukee brace non-operative treatment inidiopathic scoliosis. Clii: Orthop 1973 :93: 38-43.

Moe JH, Valuska J. Evaluation oftreatment ofscoliosis by Harringtoninstrumentation. J Bone Joint Surg [Am] 1966:48-A : 1656-7.

Monticelli G, Ascani E, Salsano V, Salsano A. Experimental scoliosisinduced by prolonged minimal electrical stimulation in theparavertebral muscles. Ito! J Orthop Traumatol 197�:1:39-54.

Nachemson A. A long term follow-up study of non-treated scoliosis.Acta Ort/zop Seand I 968:39:466-76.

Nachemson A, Nordwell A. Effectiveness of preoperative Cotrel trac-tion for correction of idiopathic scoliosis. J Bone Joint Surg [Ani]1977:59-A: 504-8.

Nash CL Jr. Current concepts review: scoliosis bracing. J Bone JointSurg [Am] 1980:62-A:848-52.

Nilsonne U, Lundgren K-D. Long-term prognosis in idiopathicscoliosis. Aeta Orthop Scand 1968:39:456-65.

Northway RO Jr, Alexander RG, Riolo ML. A cephalometric evalua-tion of the old Milwaukee brace and the modified Milwaukeebrace in relation to the normal growing child. Am J Ort/zod1974:65: 34 1-63.

Olsen GA, Rosen H, Stoll 5, Brown G. The use of muscle stimulationfor inducing scoliotic curves: a preliminary report. C/in Orthop1975:113:198-211.

Park J, Houtkin 5, Grossman J, Levine DB. A modified brace (Prenyl)for scoliosis. Clin Orthop 1977:126:67-73.

Ponder CR, Dickson JH, Harrington PR, Erwin WD. Results ofHarrington instrumentation and fusion in the adult idiopathicscoliosis patient. J Bone Joint Surg [Am] 1975:57-A:797-80l.

Ponseti IV, Friedman B. Prognosis in idiopathic scoliosis. J Bone JointSurg [Am] 1950:32-A:38l-95.

Risser JC. Scoliosis: the application of body casts for the correction ofscoliosis. Ani Aead On/zap Surg insiruct Course Lcet 1955: 12:255-9.

Risser JC. Scoliosis: past and present. J Bone Joint Surg [A�;z] 1964:46-A: 167-99.

Risser JC, Ferguson AB. Scoliosis: its prognosis. J Bone Joint Surg1936: 18:667-70.

Risser JC, Lauder CH, Norquist DM, Craig WA. Three types of bodycasts. Am AcadOrthop Surg Instruct Course Lect 1953;10: 131-42.

Roaf R. The basic anatomy of scoliosis. J Bone Joint Surg [Br] 1966:48-B: 786-92.

Rogala EG, Drummond DS, Gurr J. Scoliosis: incidence and naturalhistory: a prospective epidemiological study. J Bone Joint Surg[Ani] 1978:60-A: 173-6.

Scammon RE. The first seriatim study of human growth. Am J P/usAnthropol 1927; 10:329-36.

Schultz AB, Galante JO. Measurement of forces exerted in thecorrection of idiopathic scoliosis using three-component dy-namometers. Exp Mccli 1969:9:419-24.

Scott JC, Morgan TH. The natural history and prognosis of infantile

idiopathic scoliosis. J Bone Joint Surg [Br] l955:37-B:400-13.

Somerville EW. Rotational lordosis: the development of the single

curve. J Bone Joint Surg [Br] 1952:34-B:42l-7.

Stone B, Beekman C, Hall V, Guess V, Brooks HL. The effect of anexercise program on change in curve in adolescents with minimalidiopathic scoliosis: a preliminary study. P/irs Ther 1979:59:759-63.

Tanner JM. Grout/i at adolescence %tith a general consideration of thecijects of hereditary and environmental factors upon growth andmaturation Irons birth to nuaturitr. 2nd edn. Oxford: BlackwellScientific Publications. 1962.

Thompson SK, Bentley G. Prognosis in infantile idiopathic scoliosis. JBone Joint Surg [Br] 1980:62-B: 151-4.

Tolo VT, Gillespie R. The characteristics ofjuvenile idiopathic scoliosisand results of its treatment. J Bone Joint Surg [Br] 1978:60-B: 18 1-8.

Tupman GS. A study of bone growth in normal children and itsrelationship to skeletal maturation. J Bone Joint Surg [Br]1962:44-B:42-67.

Watts HG, Hall JE, Stanish W. The Boston brace system for thetreatment of low thoracic and lumbar scoliosis by the use of agirdle without superstructure. Cliii Orthop 1977:126:87-92.

Weinstein SL, Ponseti IV. Curve progression in idiopathic scoliosis. J

Bone Joint Surg [Am] I 983 :65-A :447-55.

Winter RB, Lovell WW, Moe JH. Excessive thoracic lordosis and lossofpulmonary function in patients with idiopathic scoliosis. J BoneJoint Surg [Am] I 975 :57-A : 972-7.

Winter RB, Carlson JM. Modern orthotics for spinal deformities. C/inOrthop 1977: 126: 74-86.

Wynne-Davies R. Infantile idiopathic scoliosis: ca usative factors,particularly in the first six months of life. J Bone Joint Surg [Br]1975:57-B: 138-41.